The dramatic re-emergence of tuberculosis has rendered present diagnostic methods inadequate. The PPD skin test detects infection by Mycobacterium tuberculosis, so infection and active disease are not distinguished. Additionally, the PPD test can often produce unreliable results: immunocompromised patients with active tuberculosis are frequently PPD-; healthy, BCG vaccinated people are PPD+, and exposure to non-pathogenic mycobacteria produce a positive PPD result. Tests that diagnose active tuberculosis ultimately depend on recovery of M. tuberculosis in culture and thus are extremely slow. Microscopic examination of stained sputum specimens remains the most rapid method of detecting active tuberculosis. However, this method often produces false negative, or equivocal results. It is thus clear that new diagnostics are necessary. Serology constitutes an attractive alternative technique for tuberculosis diagnostics, because it does not require the presence of infecting organisms in the clinical specimens, is rapid, simple and inexpensive. Lack of success in establishing a sensitive and specific ELISA for tuberculosis to this date is primarily due to limited availability of pure antigens and insufficient knowledge of humoral immune response to tuberculosis in humans. Our overall goal is to identify M. tuberculosis-specific antigens that are recognized by human immune sera and to obtain them in a readily purifiable form, in order to develop ELISA-based serodiagnosis of tuberculosis. Antigens will be identified by screening an M. tuberculosis recombinant phage expression library with human sera from various individuals (HIV+, HIV-, latent infection, active disease, during anti- tubercular therapy, different age groups). Positive clones will be characterized by restriction enzyme analysis and nucleotide sequencing, and on the basis of their reactivity with monoclonal antibodies directed against known mycobacterial antigens. Screening selected clones with M. tuberculosis DNA isolated by genomic substraction with BCG DNA will help identify antigens that are unique to M. tuberculosis. Analysis of selected antigen-expressing clones will establish a panel of antigens that will be used for screening with large numbers of sera by ELISA to begin to establish their diagnostic value. DNA fragments containing full-length genes will be cloned in recombinant gene expression vectors that allow high-level expression as well as simple one-step protein purification. During the period of requested funding, we expect to identify antigens with different specificities: antigens that specifically react with sera from infected versus diseased individuals, that detect infection/or disease in children, that are recognized by individuals coinfected with M. tuberculosis and HIV, that are recognized by antibodies that wane during successful antitubercular treatment, and that are recognized by antibodies that are maintained at detectable serum levels well past completion of therapy.